JP5782863B2 - Method for improving the performance of a stylus profilometer for surface shape measurement, and stylus profilometer for surface shape measurement using the method - Google Patents

Description

  The present invention relates to a method for improving the performance of a stylus type step meter for measuring the surface shape of a sample and a stylus type step meter for measuring the surface shape in which the method is carried out. The present invention relates to a method for improving the sensitivity and displacement resolution of a stylus type step meter composed of a probe and a displacement sensor.

  In the present specification, the term “surface shape of the sample” is meant to include the concept of the step, film thickness, and surface roughness of the sample.

  An example of a stylus step meter proposed previously to which the present invention is applied is shown in FIGS. 1 to 5 of the accompanying drawings. The stylus step meter has a rod-shaped first support member 1, The first support member 1 includes a fulcrum needle attachment member 2 extending in the left and right lateral directions at an intermediate portion thereof, and two fulcrum needles 3 are attached to both ends of the fulcrum needle attachment member 2. These two fulcrum needles 3 are supported by two fulcrum receiving members 4 (FIG. 5), whereby the first support member 1 is supported by the fulcrum receiving member 4 so as to be swingable. . At one end of the first support member 1, a measuring element of the displacement sensor 5, that is, a core 6 is attached. The displacement sensor 5 comprises a differential transformer that generates an electrical signal in accordance with the vertical displacement of the probe, and includes a coil 7.

 The other end of the first support member 1 is provided with a core 9 of a needle pressure generator 8 that applies a needle pressure to the probe, and the needle pressure generator 8 includes a coil 10. The core 9 is made of a high magnetic permeability member arranged at a position shifted in the axial direction from the center of the coil 10. A holder 12 having two magnets 11 embedded in the lower surface of the first support member 1 is centered on a line connecting the two fulcrum needles 3 at both ends of the fulcrum needle mounting member 2 of the first support member 1. Installed. As shown in FIG. 3, the holder 12 is provided with a longitudinal groove 13 having a trapezoidal cross section, and both side walls of the longitudinal groove 13 are tapered downward to form an inclined surface with respect to a horizontal plane. Yes. The two magnets 11 embedded in the holder 12 are arranged so that the polarities are opposite to each other as shown in FIG. The holder 12 containing the two magnets 11 is made of carbon in order to reduce the weight.

  1, 2, and 4, reference numeral 14 denotes a rod-like second support member. A probe 15 is attached to the tip of the second support member, and the other end is formed of a high permeability member 16. Guide protrusions 17 extending upward are formed at both ends in the longitudinal direction of the high magnetic permeability member 16, and opposing side surfaces of these guide protrusions 17 are formed as inclined surfaces that open upward. The inclined surfaces of the high permeability member 16 together with the inclined surfaces of the both side walls of the longitudinal groove 13 in the holder 12 ensure the mutual positioning when the second support member is attached to the first support member and also serve as a guide. Plays. The high permeability member 16 at the other end of the second support member 14 is fitted in the groove 13 of the holder 12 in the first support member 1, and at that time, the high permeability at the other end of the second support member 14. The member 16 is configured to contact the bottom surface of the groove of the holder 12 and not to contact the two magnets 11. Further, the groove and the high permeability member are provided with inclined surfaces as shown in FIG. 3 to ensure the mutual positioning and serve as a guide when mounting.

  Further, as shown in FIG. 3, a plate-like member 18 is provided below the high-permeability member 16 in the second support member 14, and this plate-like member 18 has a high permeability material in order to enhance the magnetic field shielding effect. The plate-like member 18 allows the replacement part to be placed in the correct position even if it is tilted and brought close to the groove 13 of the holder 12 in the first support member 1. FIG. 5 shows an enlarged structure of the fulcrum receiving member 4 that receives the fulcrum needle 3. As shown in the drawing, the fulcrum receiving member 4 has a concave surface 4a for receiving the fulcrum needle 3. The concave surface is formed in an inverted conical shape so that the fulcrum needle 3 is positioned and received with high accuracy.

  The illustrated stylus profilometer constructed in this way is provided with a displacement sensor 5 and a needle pressure generating device 8 at both ends, and is supported by two fulcrum receiving members 4 via a fulcrum needle 3 so as to be swingable. A second support member 14 having a probe 15 and a high magnetic permeability member 16 at both ends is fixed to the holder 12 of the first support member 1 by the magnet's attracting force. In this case, the second support member 14 is formed by the inclined surfaces of both side walls of the longitudinal groove 13 in the holder 12 and the opposed inclined surfaces of the guide protrusions 17 in the high permeability member 16 of the second support member 14. The support member 1 can be easily positioned by being accurately positioned at a predetermined position with respect to the holder 12. Then, by applying a predetermined current to the coil 10 of the needle pressure generator 8, a force is generated according to the magnitude of the current, and the core 9 of the needle pressure generator 8 is drawn into the center of the coil 10 by this force. It is. As a result, the first and second support members 1 and 14 swing via the fulcrum needle 3 and press the probe 15 against the sample. By scanning the sample or the detection system, the probe 15 traces the surface of the sample, and the first and second support members 1 and 14 rotate slightly around the fixed fulcrum according to the surface shape. The displacement of the core 6 of the differential transformer 5 is detected, and the displacement of the core 6 is measured with high accuracy with low noise by a digital lock-in amplifier using a digital signal processor DSP, for example. The surface shape and level difference of the sample are measured by converting to displacement.

JP 2006-226964 A JP2009-20050

 In the previously proposed structure, the distance between the fulcrum and the probe is increased in order to reduce the needle jump at the step portion of the sample, but the sensitivity of such a stylus type step gauge is the sensitivity of the displacement sensor. Is multiplied by “distance between fulcrum and displacement sensor core / distance between fulcrum and probe”. Here, the sensitivity is an output voltage per unit displacement. The ratio of the distance is 0.5, for example, and the sensitivity is 0.5 times. This is a design that, for example, 0.03 mgf is a very small force in this kind of step gauge, and makes it difficult for the needle to fly even when measuring easily deformable samples. Yes.

 However, the force range that is frequently used in practice is as large as several mgf to 10 mgf. In this case, the probe is difficult to skip, so the distance between the fulcrum and the probe may be reduced. In addition, “sensitivity of the stylus type step meter” and “proneness of the probe” are contradictory, but it should be designed to reduce the needle jump even if the sensitivity of the stylus type step meter is the same. is there.

 Therefore, the problem to be solved by the present invention is to improve the sensitivity and displacement resolution of the step gauge as much as possible in the practical force region of 1 mgf or more without any needle jump.

In order to solve the above-described problems , according to the method for improving the surface shape measuring stylus profilometer according to the first aspect of the present invention , a probe is provided at one end of a support body swingably attached to a fulcrum. Mount the needle, attach the magnetic core of the displacement sensor that detects the vertical displacement of the probe to the other end of the support, and use the displacement sensor to change the surface shape of the sample captured by the probe by rotating around the support fulcrum. In the method for improving the performance of the stylus profilometer for measuring the surface profile,
F is the force applied to the probe, I is the moment of inertia about the fulcrum, r is the distance between the fulcrum and the probe, z is the displacement of the probe tip, and t is the time. Equation F = I / r ² d ² z / dt ²
Based on
At least the force F applied to the probe is set to 1 mgf or more, and the distance r between the fulcrum and the probe is determined according to the force applied to the probe,
The tip of the probe is lowered with respect to the fulcrum so that the line connecting the fulcrum and the tip of the probe is inclined 45 degrees below the horizontal line between the fulcrum and the probe while the support is horizontal. It is said.

Further , according to the stylus profilometer for surface shape measurement according to the second aspect of the present invention , a probe is attached to one end of a support that is swingably attached to a fulcrum, and a probe is attached to the other end of the support. A stylus profilometer for measuring the surface shape of a sample that is attached to a magnetic core of a displacement sensor that detects the displacement of the needle in the vertical direction and that measures the surface shape of the sample captured by the probe using a displacement sensor by means of a rotational movement around the fulcrum of the support. In
F is the force applied to the probe, I is the moment of inertia about the fulcrum, r is the distance between the fulcrum and the probe, z is the displacement of the probe tip, and t is the time. Equation F = I / r ² d ² z / dt ²
Based on
At least the force F applied to the probe is 1 mgf or more, and the distance r between the fulcrum and the probe is determined according to the force applied to the probe,
The tip of the probe is lowered with respect to the fulcrum so that the line connecting the fulcrum and the tip of the probe in a state where the support is level is inclined 45 degrees below the horizontal line between the fulcrum and the probe. It is characterized by that.

Regarding the force F applied to the probe, for example, when a force 16 times the conventional force F is used as the force F applied to the probe, the distance r between the fulcrum and the probe may be ¼ (I is Is not going to change). At that time, the equation of motion does not change (both sides are 16 times larger and d ² z / dt ² does not change), so the motion z (t) of the tip of the probe does not change. That is, the “proneness of the probe” and the “size of the needle” do not change. As a specific example, if it is used at F = 0.1 mgf, if the step meter with r = 40 mm may be increased to F = 1.6 mgf, the “needle skipping difficulty” can be increased even if r = 10 mm. The same. In this case, since r is 1/4 times, the sensitivity as a step meter is 4 times.

  If the horizontal distance between the fulcrum and the probe is the same when the line connecting the fulcrum and the tip of the probe is tilted 45 degrees from the horizontal, the geometrical considerations indicate that the line connecting the fulcrum and the tip of the probe is horizontal. The sensitivity is the same as in the case of. However, the distance from the fulcrum to the tip of the probe is twice as large as the route, and when the equations of motion of the tip of the probe are compared in these two cases, the size of the probe jump is 0.707, which is the reciprocal of route 2. You can see that it doubles. Therefore, even if the sensitivity of the level difference meter is the same, the needle jump is reduced to 0.707 times. It can be seen that when the force is the same, the needle jump size is 0.707 times, and the force at which the jump size is the same is 0.707 times smaller. That is, the force F may be 0.707 times, and the previous example (“F = 0.1 mgf, r = 40 mm” → “F = 1.6 mgf, r = 10 mm, the above connecting line is horizontal”) Thus, when the above line is set at 45 degrees from the horizontal, the force F is 16 times x 0.707 = 11.3 times with respect to the needle jump, and the horizontal distance can be offset by 1/4. Accordingly, “F = 1.1 mgf, horizontal distance 10 mm, 45 degrees” and “probing difficulty of the probe” are the same as the previous example “F = 1.6 mgf, r = 10 mm, horizontal”. In addition, when the line connecting the fulcrum and the tip of the probe is inclined 45 degrees from the horizontal and the tip of the probe is lowered, the probe moves in the scanning direction when the probe rises at the step portion during scanning. Ascends in the opposite direction. Therefore, the rising speed becomes small. That is, the initial velocity in the z direction is reduced. As a result, the needle jump height and jump time are reduced.

  According to the present invention, it is possible to provide a step meter having a high sensitivity in a practical force range of 1 mgf or more, and “the arrangement of the fulcrum and the probe” that reduces the needle skip while maintaining the same sensitivity. It is possible to provide a level difference meter that has a sensitivity four times that of the prior art and has no needle skipping. Displacement noise is reduced by improving sensitivity, and the “peak-to-peak” noise is reduced to 0.16 nm at the atomic level in actual measurement, and displacement resolution is improved. Therefore, a differential transformer is used as a displacement sensor, and an inexpensive step gauge with a very simple structure can obtain a high displacement resolution at the atomic level and measure a surface shape that was previously hidden by noise and could not be seen. become.

  At a practical probe force range of 1 mgf or more, the sensitivity can be improved and the displacement resolution can be improved to 0.16 nm without causing the probe to skip at the stepped portion of the sample. Further, by increasing the ratio of the distance between the fulcrum-probe and the distance between the fulcrum-sensor, the sensitivity as a step meter can be improved. Further, by tilting the line connecting the fulcrum and the tip of the probe 45 degrees from the horizontal and lowering the probe tip, the “force for preventing the probe from jumping” is reduced to 0.707 without decreasing the sensitivity improvement rate. As a result, the needle jump can be suppressed even with a weak force, and the vertical rising speed of the tip of the probe at the stepped portion can be reduced, thereby suppressing the size of the needle jump. It becomes like this.

The schematic of the structure of the stylus type level difference meter which is implementing the method of this invention. The schematic diagram which looked at the principal part of the stylus type level difference meter in FIG. 1 from the bottom. The expanded partial sectional view along the AB line which shows the structure of the holder part of the stylus type level difference meter in FIG. The figure which looked at the holder part of the stylus type level difference meter in FIG. 1 from the bottom. The expanded sectional view which shows the structure of the fulcrum part of the stylus type level difference meter in FIG. The figure which shows the example of arrangement | positioning of a probe, a fulcrum, and a displacement sensor core. The figure which shows another example of arrangement | positioning of a probe, a fulcrum, and a displacement sensor core. The figure which shows the example of arrangement | positioning of a probe, a fulcrum, and a displacement sensor core when a probe is below. The figure which shows another example of arrangement | positioning of a probe, a fulcrum, and a displacement sensor core in case a probe is below. The diagram showing the arrangement of the probe and the movement of the probe around the fulcrum. The graph which shows the example of a measurement of the displacement noise when not scanning a sample. The figure which shows the example which measured displacement noise repeatedly. The graph which shows the measurement example of the voltage noise density of a sensor output. The graph which shows the example which rounded the measurement data to the discrete value which divided | segmented the measurable range by 16 bits. The graph which shows the example which scanned the level | step-difference part of the sample by 0.1 mgf. The graph which shows the example which scanned the level | step-difference part of the sample by 2.1 mgf. The graph which shows the example which scanned the same place on a sample 3 times continuously by 1.6 mgf.

Embodiments of the present invention will be described below.
In the structure example of the stylus profilometer shown in FIG. 1, F is the force applied to the tip of the probe, z is the z-direction position of the tip of the probe, I is the moment of inertia around the fulcrum, and the tip of the probe is from the fulcrum. If the equation of motion around the fulcrum is transformed with the distance up to r, the following equation is obtained (see, for example, Patent Document 1).

F = I / r ² d ² z / dt ² (1)

This is because the force F works and the mass is I / r ² It can be regarded as a movement of mass points. While the probe is stopped, a constant force is generated by the force generating coil, so it can be regarded as the free fall motion of the mass point in a constant gravitational field. If F is constant, d ² z / dt ^{2 is} also constant. In such a case, it is assumed that the initial velocity in the z direction of the tip of the probe (the speed at which the tip of the probe moves away from the sample surface) is v ₀ and that the center of gravity of the movable part supported by the fulcrum is close to the fulcrum. The height h of the jump and the time when the tip of the probe is broken (that is, the time until the tip of the probe leaves the sample surface and then returns to the surface of the same height) 2t ₀ are respectively expressed by the following equations: Represented.

_{^{h = Iv 0 2/2 r}} 2 F (2)

2t ₀ = 2 I v ₀ / r ² F (3)

  That is, the larger the r, the smaller the needle jump. However, as can be seen from FIG. 6, if the distance between the fulcrum and the probe is large (the number in the figure indicates the ratio of the length), the change in the probe displacement z is reduced and transmitted to the displacement sensor core. Sensitivity as a total is reduced. In a design that assumes measurement with a small force up to 0.03 mgf, r is, for example, 40 mm, and the distance between the fulcrum and the displacement sensor core is 20 mm (the smaller I is, the smaller the needle jump is, so the displacement sensor core is at the fulcrum. The closer one is better for needle skipping). Therefore, the sensitivity of the step gauge is half that of the displacement sensor. This is because of the design that prioritizes needle jump over sensitivity.

Assuming only the measurement at a practical force of about 1 mgf or more, as can be seen from the equations (1) to (3), r can be reduced as F is larger. Assuming that I does not change, the needle jump size does not change if r ² F is constant. Therefore, r can be determined from the relationship of r Ｆ F ^−0.5 . For example, if F is 16 times, r may be 1/4 times (FIG. 7).

  If the moment of inertia I around the fulcrum is the same, “F = 0.1 mgf, r = 40 mm” and “F = 1.6 mgf, r = 10 mm” are the same with respect to needle skipping. And the sensitivity as a level difference meter becomes 4 times.

Regarding the force F required for the probe not to stop at the step portion of the sample, I / r ² = 0.114 g for the step meter of r = 40 mm, and the curvature radius of the tip of the probe is 2.5 μm. In this case, the force required for the needle not to stop is about 0.1 mgf (0.08 mgf to 0.18 mgf depending on the step) at a scanning speed of 100 μm / s. When the scanning speed is 50 μm / s, which is half, the force F is ¼.

  In this step meter, when the distance r from the fulcrum to the tip of the probe is set to 1/4, 10 mm, the needle jump does not occur if the force F is increased 16 times. In other words, needle skipping does not occur at about 1.6 mgf or more. This is a practical power range that is often used in practice.

  Based on the above consideration based on the formula (1), a step gauge can be designed in which the needle skip does not occur at a practical force range of about 1.6 mgf or more and the sensitivity is four times that of the conventional one.

  As shown in FIGS. 8 and 9, the probe 15 is provided in an actual step gauge so that measurement can be performed at an arbitrary place on a flat and wide sample such as a silicon wafer for a semiconductor element or a glass substrate for a liquid crystal display. A box-shaped sensor head, which is positioned at the bottom of the step meter components and accommodates the step meter components of FIGS. 8 and 9, and only the tip of the probe 15 protrudes downward, moves on the sample. Alternatively, the sensor head may be fixed so that the sample moves. Therefore, the line connecting the fulcrum and the probe is inclined from the horizontal as shown in FIGS. In the structure shown in FIG. 8, the horizontal distance between the fulcrum and the probe is 40 mm, and in the structure shown in FIG. 9, the horizontal distance is 10 mm. The numbers 2 and 4 in FIG. 8 and the numbers 1 and 2 in FIG. 9 represent the length ratio.

  In FIG. 8, the “horizontal distance between the fulcrum and the probe” is 0.97 times the “distance r between the fulcrum and the tip of the probe”, Δz described later is 0.97 times Δl, and Δz and Δl are Since they are substantially equal, the movement of the probe 15 is substantially the same as when the “line connecting the fulcrum and the probe 15” is horizontal.

FIG. 10 shows the movement of the probe around the fulcrum. Consider the case where the tip of the probe is at points A, B, and C, respectively. The equation of motion around the fulcrum of the probe tip is given by the following equation with θ as the rotation angle (θ and F are clockwise in FIG. 10).

r F = I d ² θ / dt ² (4)

From Δθ = Δl / r, where l is the tangential displacement, F = I / r ² d ² l / dt ² (5)

When the tip of the probe is point A, from Δl = Δz, F = I / r ² d ² z / dt ² (6)

At the point B of the probe, Δl = 1.414 Δz F = 1.414 I / r ² d ² z / dt ² (7)

When the tip of the probe is point C, I is the same, and when the distance to the fulcrum is r _c and the force there is F
F = I / r _c ² d ² z / dt ²
From r _c = r / 1.414 F = 2I / r ² d ² z / dt ² (8)

It becomes. From the comparison between Equation (7) and Equation (8), at point B, the same z acceleration is obtained with a force 0.707 times that at point C. Therefore, it can be seen that the point B has the same effect with a weak force 0.707 times that of the point C as an effect of suppressing the needle jump.

  From the angle of rotation with respect to the change Δz in the height of the sample surface or the geometrical consideration of Δl, it can be seen that the sensitivity is the same at points B and C. The change in the z direction of the tip of the probe accompanying the change in the height of the sample surface is naturally the same at point B and point C. That is, the point B is more advantageous than the point C because the needle jump can be suppressed with 0.707 times weaker force while maintaining the same sensitivity. In addition, at point B, the force is applied to the sample with an inclination of 45 degrees, but there is no particular problem.

  From the above, in the arrangement of the fulcrum and the probe 15 shown in FIG. 9, the sensitivity remains the same as that shown in FIG. 7, and the same amount of needle jump suppression effect can be obtained even with 0.707 times weaker force. . In the above discussion, in FIG. 7, it is assumed that needle skip does not occur at 1.6 mgf or more. However, in this arrangement, needle jump does not occur at 1.6 mgf × 0.707 = 1.1 mgf or more.

  When scanning the sensor head to the right in FIG. 10, considering that the tip of the probe rises from the lower stage to the upper stage of the stepped portion of the sample, the tip of the probe is at point B compared to point A or point C. Since z increases while moving to the left in FIG. 10, the time until completion of the rise at the step portion of the sample becomes longer. Therefore, Δz / Δt becomes small and the rising speed becomes small. As a result, the initial velocity in the z direction is reduced, and the needle jump is reduced as can be seen from equations (2) and (3). Thus, when the “line connecting the fulcrum and the probe” is lowered by 45 degrees from the horizontal, for example, the initial velocity in the z direction when the probe moves up the step is reduced, and the needle jump is reduced.

  When the probe moves up the step, the probe moves in the direction opposite to the scanning direction, so that the rise at the step portion of the measurement data is slightly dull. For example, if the step is 1 μm, the scan distance required for the rise is only 1 μm longer. So there is no problem in practical use.

  An example of displacement noise measured with the arrangement shown in FIG. 9 is shown in FIG. The horizontal distance between the fulcrum and the probe is 10 mm, the distance between the fulcrum and the displacement sensor core is 20 mm, and the line connecting the fulcrum and the probe is 45 degrees from the horizontal. The probe is lowered onto the sample, the force applied to the probe is 1.7 mgf, and the probe displacement z is measured for 4 seconds every 1 ms in a stationary state without scanning. The primary voltage of the differential transformer is 5 kHz, the secondary voltage is measured by a digital lock-in amplifier configured using a digital signal processor (DSP), the displacement of the displacement sensor core is measured, and the displacement of the probe tip is calculated. ing. The measured data is subjected to a low-pass filter LPF (finite impulse response FIR, order 90, using Blackman window function) having a cutoff frequency fc = 15 Hz.

  The probe sensitivity increased 3.93 times compared to the configuration of FIG. As a result, the displacement noise was reduced to about ¼ (details will be described later), and the “peak-peak” of the noise was much smaller than 0.2 nm as in the example shown in FIG. FIG. 11 also shows the swell equivalent of 0.5 Hz or less included in the measurement data. FIG. 12 shows the result of measuring “peak-peak” in 1 second after removing such a corrugated portion more than 100 times every 6 seconds. This is the measurement result when the force applied to the probe is 0.18 mgf, the average value is 0.163 nm, the standard deviation of the data in 1 second is 0.034 nm, and the arithmetic average roughness in 2 seconds The Ra equivalent was 0.028 nm, which was a very small value. When the force applied to the probe is 0.85 mgf, the “peak-peak” in 1 second is 0.162 nm, and when the force applied to the probe is 1.50 mgf, The “peak-peak” was 0.160 nm.

  In the configuration of FIG. 8, the “peak-peak” of noise for 1 second was 0.62 nm, but it was 0.62 nm / 3.93 = 0.158 nm. It can be seen that this appears as a decrease in.

  In a stylus type step meter for measuring a level difference in a thin film process on a normal silicon substrate or glass substrate, the “peak-peak” in 1 second of noise as described above is about 1 to 2 nm. It can be said that the measurement result in FIG. Therefore, the displacement resolution is about one digit higher.

FIG. 13 shows the result of measuring the voltage noise density in terms of input to the preamplifier in a stationary state where the probe is lowered on the sample using the arrangement configuration of FIG. 9 (Hanning in the route of the power spectral density PSD). Use windows). This is the noise density of the output voltage of the displacement sensor. Although it is conversion from the measurement data before LPF of f _c = 15 Hz, it relates to the data that has been subjected to the moving average processing for 230 Hz. An instrumentation amplifier having an input conversion voltage noise density of 1 nV / Hz ^0.5 is used as a preamplifier between the displacement sensor and the measuring instrument. The main noise sources are the thermal noise of the resistance of the secondary coil and the Barkhausen noise of the displacement sensor core. The magnitude of the voltage noise density is the same as that of the displacement sensor alone. The sensitivity increases with the ratio of “distance between fulcrum-displacement sensor core” / “distance between fulcrum-probe”, and the displacement is the reciprocal thereof. It can be seen that the noise is reduced.

  FIG. 14 shows the time change of the needle tip displacement in a stationary state with the needle lowered as in FIG. 11, but the measurable range is divided by 16 bits (65536), and the measurement data is rounded to its discrete value. -An example of transfer to a computer PC at 232C is shown. In this example, the measurable range is divided from −833.8 nm to +833.8 nm by 16 bits, and becomes a discrete value every 0.0254 nm. Note that such a “discrete value interval” that is merely a resolution on display may be referred to as resolution, but it does not represent the measurement performance of the displacement measuring apparatus, and is not the original displacement resolution.

  FIG. 15 shows an example in which a needle jumps. FIG. 10 shows a measurement result obtained by scanning a step sample having a force of 0.1 mgf applied to the probe and about 2 μm at 100 μm / s using the arrangement configuration of FIG. As described above, it is known from the calculation that the needle jumps when the probe arrangement, the value of the moment of inertia I, and the scanning speed are less than 1 mgf. Since the force is an order of magnitude less than the force to stop it, the needle is sharply broken.

  In FIG. 16, the result of having measured the same level | step difference by 2.1 mgf is shown. The same result was obtained with 1.4 mgf, and the needle was not skipped. That is, the result as the above-mentioned consideration was confirmed by actual measurement.

  In the force range of 1 mgf or more, which is often used practically, there is no needle skipping, the sensitivity is 4 times that of the conventional one, and the displacement resolution (peak-to-peak in 1 second of noise) is 0.16 nm and the level difference at the atomic level A total was made.

FIG. 17 shows the measurement results of scanning the same place on the sample three times at 71 μm / s with the force of 1.6 mgf applied to the probe using the arrangement configuration of FIG. The LPF is processed with a Bessel characteristic first order at f _c = 15 Hz. The least square regression curve quartic polynomial is subtracted from the first measurement data, and the same polynomial is subtracted from the second measurement data and the constant is added. The thin line is the same polynomial from the third measurement data. The dotted line adds another constant.

  In order to examine the reproducibility of the measurement result, the above-mentioned constant value was subtracted from the measurement value. The above constants are determined so that the values on the vertical axis in FIG. FIG. 17 shows that the measurement results are reproduced in the range of about 0.1 to 0.2 nm, apart from the constants. From this, it was confirmed that the surface shape in the z direction can be measured with an atomic level displacement resolution of about 0.1 to 0.2 nm.

1: first support member 2: fulcrum needle mounting member 3: fulcrum needle 4: fulcrum receiving member 5: displacement sensor 6: core 7: coil 8: needle pressure generator 9: core 10: needle pressure generator 8 Coil 15: probe

Claims (2)

Attach a probe to one end of a support that is swingably attached to a fulcrum,
Attach the magnetic core of the displacement sensor that detects the vertical displacement of the probe to the other end of the support,
In the method for improving the performance of a stylus-type step gauge for measuring the surface shape, which measures the surface shape of the sample captured by the probe with a displacement sensor by the rotational movement around the support point of
F is the force applied to the probe, I is the moment of inertia about the fulcrum, r is the distance between the fulcrum and the probe, z is the displacement of the probe tip, and t is the time. Equation F = I / r ² d ² z / dt ²
Based on
At least force F applied to the probe is 1 mgf or more,
The distance r between the fulcrum and the probe is determined according to the force applied to the probe,
Lowering the tip of the probe downward with respect to the fulcrum so that the line connecting the fulcrum and the tip of the probe with the support level is inclined 45 degrees below the horizontal line between the fulcrum and the probe <br> A method for improving the performance of a stylus profilometer for measuring the surface shape. A probe is attached to one end of a support that is swingably attached to a fulcrum, and a magnetic core of a displacement sensor that detects the vertical displacement of the probe is attached to the other end of the support. In a stylus type step meter for measuring the surface shape, which measures the surface shape with a displacement sensor by rotational movement around the fulcrum of the support,
F is the force applied to the probe, I is the moment of inertia about the fulcrum, r is the distance between the fulcrum and the probe, z is the displacement of the probe tip, and t is the time. Equation F = I / r ² d ² z / dt ²
Based on
At least the force F applied to the probe is 1 mgf or more, and the distance r between the fulcrum and the probe is determined according to the force applied to the probe,
The tip of the probe is lowered with respect to the fulcrum so that the line connecting the fulcrum and the tip of the probe in a state where the support is level is inclined 45 degrees below the horizontal line between the fulcrum and the probe. A stylus-type step gauge for measuring surface shape.

Images